Emc IP Holding Company LLC

Dell Products Inc.

Emc Corporation

Dell & Products L.p.

Adriana Bechara Prado

Paulo Abelha Ferreira

Pablo Nascimento Da Silva

Jonas Furtado Dias

Tiago Salviano Calmon

André De Almeida Maximo

Rodrigo Dias Arruda Senra

Rômulo Teixeira De Abreu Pinho

Angelo E M Ciarlini

Vinícius Michel Gottin

Victor Bursztyn

Vinicius Michel Gottin

Karin Breitman

Vitor Silva Sousa

Owen Martin

Adnan Sahin

Ron Arnan

Ravi Shukla

Eduardo Vera Sousa

Paulo De Figueiredo Pires

Roberto Nery Stelling Neto

Jeffrey Scott Vah

Werner Spolidoro Freund

Diego Salomone Bruno

Jaumir Valença Da Silveira Junior

Fábio André Machado Porto

Hagay Dagan

Iam Palatnik De Sousa

Wagner Dos Santos Vieira

Aaron Sanchez

Isabella Costa Maia

Karen Stéfany Martins

Monica Barros

David Burth Kurka

Marcia Lucas Pesce

Eduarda Tatiane Chagas

Alexander Eulalio Robles Robles

Thais Luca Marques De Almeida

Altobelli De Brito Mantuan

Rodolpho Rosa Da Silva

Rodrigo De Souza Lima Espinha

The disclosure relates to information retrieval and includes a preprocessing stage and a retrieval stage. The preprocessing stage includes ingesting data in a semantic manner that incorporates domain expert knowledge and that retains contextual information. The ingested data or objects are chunked and chunk embeddings are generated. A knowledge graph is created to hierarchically represent each of the objects. During the retrieval stage, the top k embeddings closest to an embedding of a query are used to identify the semantic cells and metadata cells of a document corresponding to the top k embeddings. This generates a return object that is used an input for content generation.

Semantic segmentation of data for information retrieval

Saliency heatmap generation is disclosed. A mask that corresponds to an image is iteratively passed through a model. At each iteration, the output mask is optimized via gradient descent to gradually turn off one or more pixels. The output mask is iteratively passed in order to minimize contributions of pixels and to maximize a class prediction. After generating a final output mask, a saliency heatmap is generated.

Detailed perturbative heatmap explanations

One example method may be performed in a distributed environment that includes a group of nodes configured to communicate with each other, and to communicate with a central node of the distributed environment. The example method includes performing, by two or more of the nodes, the operations of generating a public/private key pair, generating a secret shared key using the respective public keys of the other nodes, using the secret shared key to generate isometric transformations of data of the node, and transmitting the transformed data to the central node.

Non-anonymized privacy preserving global and local anomaly detection in distributed systems

One example method includes accessing input data elements from logs that identify user problems with computing system components, the data elements each associated with a respective original class label that identifies a class of computing system components to which the data element relates, the respective original class labels forming a group of class labels, and a first of the original class labels is overrepresented in the group, and reducing overrepresentation of the first original class label in the group by creating an arbitrary aggregation of some of the class labels that includes the first original class label. The method includes creating, based on a hierarchical modeling structure, prepared data in which an original class label is replaced by the aggregation. Next a hierarchical model and benchmark model are trained, and each model generates respective predictions for comparison. An inferencing process is performed to determine which predicted label will be used.

Hierarchical modeling approach for digital repair parts prediction

Decoding random forest problem solving through node labeling and subtree distributions. Random forests, like any other type of machine learning algorithm, are designed and configured to solve classification, regression, and/or prediction problems. Solutions (or outputs) provided by random forests, given inputs in the form of values for a set of features, may sometimes be inaccurate, unexpected, or undesirable. Understanding or decoding how a random forest solves a given problem may be a way to correct or improve the random forest. The disclosed method, accordingly, proposes decoding random forest problem solving through the identification of subtrees (by way of node labeling) amongst a random forest, as well as the frequencies that these subtrees appear (or distributions thereof) throughout the random forest.

Decoding random forest problem solving through node labeling and subtree distributions

Training examples are created from telemetry data, in which each training example engineered features derived from the telemetry data, storage system characteristics about the storage system that processed the workload associated with the telemetry data, and the response time of the storage system while processing the workload. The training examples are provided to an unsupervised learning process which assigns the training examples to clusters. Training examples of each cluster are used to train/test a separate supervised learning process for the cluster, to cause each supervised learning process to learn a regression between independent variables (system characteristics and workload features) and a dependent variable (storage system response time). To determine a response time of a proposed storage system, the proposed workload is used to select one of the clusters, and then the trained learning process for the selected cluster is used to determine the response time of the proposed storage system.

Workload-oriented prediction of response times of storage systems

A global prediction manager for generating predictions using data from data zones includes storage for storing a model repository comprising a global model set and a prediction manager. The prediction manager obtains a local model set from a data zone of the data zones indicating that the global model set is unacceptable; makes a determination that the local model set is acceptable; in response to the determination: distributes the local model set to at least one second data zone of the data zones; obtains compressed telemetry data, that was compressed using the local model set, from the data zone and the at least one second data zone; and generates a global prediction regarding a future operating condition of the data zones using: the compressed local telemetry data and the local model set.

Locality-aware compressor-decompressor for keeping prediction models up-to-date in resource constrained networks

Techniques are provided for allocation of shared computing resources using source code feature extraction and machine learning techniques. An exemplary method comprises obtaining source code for execution in a shared computing environment; extracting a plurality of discriminative features from the source code; obtaining a trained machine learning model; and generating a prediction of an allocation of one or more resources of the shared computing environment needed to satisfy one or more service level agreement requirements for the source code. The generated prediction is optionally adjusted using a statistical analysis of an error curve, based on one or more error boundaries obtained by the trained machine learning model. The trained machine learning model can be trained using a set of discriminative features extracted from training source code and corresponding measurements of metrics of the service level agreement requirements obtained by executing the training source code on a plurality of the resources of the shared computing environment.

Allocation of shared computing resources using source code feature extraction and machine learning

An autoregressor that compresses input data for a specific purpose. Input data is compressed using a compression/decompression framework and by accounting for a purpose of a prediction model. The compression aspect of the framework is distributed and the decompression aspect of the framework may be centralized. The compression/decompression framework and a machine learning prediction model can be centrally trained. The compressor is distributed to nodes such that the input data can be compressed and transmitted to a central node. The model and the compression/decompression framework are continually trained on new data. This allows for lossy compression and higher compression rates while maintaining low prediction error rates.

Compression and decompression of telemetry data for prediction models

A prediction manager for providing responsiveness predictions for deployments includes persistent storage and a predictor. The persistent storage stores training data and conditioned training data. The predictor is programmed to obtain training data based on: a configuration of at least one deployment of the deployments, and a measured responsiveness of the at least one deployment, perform a peak extraction analysis on the measured responsiveness to obtain conditioned training data, obtain a prediction model using: the training data, and a first untrained prediction model, obtain a confidence prediction model using: the conditioned training data, and a second untrained prediction model, obtain a combined prediction using: the prediction model, and the confidence prediction model, and perform, based on the combined prediction, an action set to prevent a responsiveness failure.

Confident peak-aware response time estimation by exploiting telemetry data from different system configurations

Techniques are provided for provenance-based software script reuse. One method comprises extracting provenance data from source code including, for example, source code fragments, wherein the extracted provenance data indicates a control flow and a data flow of the source code; encapsulating source code fragments from the source code that satisfy one or more similarity criteria as a reusable source code fragment; and providing a repository of encapsulated reusable source code fragments for reuse during a development of new software scripts. The repository of encapsulated reusable source code fragments optionally comprises a searchable database further including, for example, the provenance data, data annotations, input parameters and generated results for the corresponding source code fragment.

Provenance-based reuse of software code

A deployment manager includes storage for storing a prediction model based on telemetry data from the deployments and a prediction manager. The prediction manager generates, using the prediction model and second telemetry data obtained from a deployment of the deployments: a prediction, and a prediction error estimate; in response to a determination that the prediction indicates a negative impact on the deployment: generates a confidence estimation for the prediction based on a variability of the second telemetry data from the telemetry data; in response to a second determination that the confidence estimation indicates that the prediction error estimate is inaccurate: remediates the prediction based on the variability to obtain an updated prediction; and performs an action set, based on the updated prediction, to reduce an impact of the negative impact on the deployment.

Framework for measuring telemetry data variability for confidence evaluation of a machine learning estimator

Techniques are provided for allocating shared computing resources using source code feature extraction and cluster-based training of machine learning models. An exemplary method comprises: obtaining a source code corpus with source code segments for execution in a shared computing environment; extracting discriminative features from the source code segments in the source code corpus; obtaining a trained machine learning model, wherein the trained machine learning model is trained using samples of source code segments from clusters derived from clustering the source code corpus based on (i) a term frequency metric, and/or (ii) observed values of execution metrics; and generating, using the trained machine learning model, a prediction of an allocation of resources of the shared computing environment needed to satisfy service level agreement requirements for source code to be executed in the shared computing environment. The discriminative features may be extracted from the source code corpus using natural language processing techniques and/or pattern-based techniques.

Allocation of shared computing resources using source code feature extraction and clustering-based training of machine learning models

One or more aspects of the present disclosure relate to providing storage system configuration recommendations. System configurations of one or more storage devices can be determined based on their respective collected telemetry information. Performance of storage devices having different system configurations can be predicted based on one or more of: the collected telemetry information and each of the different system configurations. In response to receiving one or more requested performance characteristics and workload conditions, one or more recommended storage device configurations can be provided for each request based on the predicted performance characteristics, the requested performance characteristics, and the workload conditions.

Storage system configuration based on workload characteristics and performance metrics

Telemetry data gathered from active deployed SAN nodes is used to create a machine learning model that predicts storage system performance, e.g. in terms of response time. The telemetry data may be filtered to remove outlier values and less relevant information before creating the training dataset. Engineered features may be created that include types of data that are not present in the telemetry data. For example, data types from the telemetry data may be combined to create engineered features that are more relevant than the individual data types. The engineered features are included in the training dataset. The machine learning model may be used to test possible configurations for a planned SAN node based on expected workload and performance requirements. Outputted data may include satisfactory configurations for a planned storage system.

Using telemetry data from different storage systems to predict response time

Growing community of inventors

Rio de Janeiro, Brazil

Adriana Bechara Prado